86 ME. G. B. H. BAEEETT-HAMILTON ON [Feb. 7, 



" The possibility of the shells having been brought as bait for 

 the lines of the fishing-boats visiting the bank is, I think, disposed 

 of by the mixed character of the deposit, some of the shells being 

 unsuitable for such a purpose. It would be interesting to trace 

 out the area occupied by these dead shells, and, possibly, to search 

 in a similar manner for the lost land of Hy Brassil on the Porcupine 

 Bank, but the time at our disposal only gave us the chance for one 

 dip into this deposit."' 



Turning to Mas imisculus we have to deal with a very different 

 species, and I do not in this case attempt to prove that this animal 

 has reached St. Kilda without the help of man. That it must have 

 existed there for a considerable time, perhaps for hundreds of years, 

 is, however, as 1 hope to show, very probable. Well known and 

 w-idely spread in almost all regions where the habitations of man 

 afford it a refuge, it is impossible to state what is the native home 

 of the species. Xot only is the domestic form of Mus muscidus 

 ■ft-idely spread and readily susceptible of introduction into the 

 houses of its unwilling protector, man, but its variability- is as 

 remarkable as is the constancy to type of Mus sylvaticus. Still it 

 has never, I believe, been asserted that the species is anything but 

 an introduction into Western Europe and the British Islands. 



Light or yellowish varieties of 2Ius mnsculus have from time to 

 time received names, such as M. hortidanus Nordmann, M. norcl- 

 manni Keys. & Blasius, 31. JIavescens Fisher, and J/. Jfaviventris 

 Lalaste ; the last two names preoccupied by other species of the 

 genus. In addition, however, to these almost domestic members 

 of the Mits-muscuhis group of Mice, we have in many parts o£ 

 the world wild forms of Mice which, though differing to a 

 greater or less extent in their size, length of tail and colour, cannot 

 be distinguished from Mus muscidus in their skull and teeth. 

 Such Mice are M. hactrianus Blyth and M. gentUis Brants, which 

 are widely distributed in the deserts respectively of Asia and 

 N. Africa, and J/, wacjneri Eversm. { = M. pachycei-eos Blanford) 

 of Central Asia, the latter a true House-Mouse, often found 

 inhabiting houses, and differing in no cranial characters from Mus 

 muscidus proper. 



Lastly we have a set of Mice also of varied colours, size of body, 

 and proportion of tail, but mostly characterized by the posses- 

 sion of a white belly, which are found in many of the regions 

 where typical Mus mnsculus occurs. Such are M. spretus Lataste, 

 of the Barbary States, and M. spicdegus Petenyi, of Hungary, 

 France, Portugal, and Western Europe. These mice may occur in 

 close proximity to the typical Mus muscidus, as was found by 

 Mr. Oldfield Thomas in Portugal and by myself in Morocco. 



Among all these perplexing forms it is indeed difficult to assign 

 a proper place to M. mundis, and more so to hazard even a guess 

 as to the possible origiu of the domestic races of Mus mnsculus. 

 We know, however, that almost wherever there are deserts 

 there a bactrianus-like Mouse is found, so that M. hactrianus is 

 perhaps as widely distributed in deserts as is Mus musculus 



